Transcript

Jen

This is The Web Ahead, a weekly conversation about changing technologies and the future of the web. I'm your host Jen Simmons and this is episode number 60. I want to say first thank you so much to today's sponsors, there are three of them today: FreshBooks, LegalZoom, and New Relic, I'll talk about those more later in the show. Hello, hello podcast listeners! It's been a while since I've done a show, it's now the end of February, 2014. Almost March. March will come in three days. Oh my, I looked to see when I did the show last and it says here November 1, 2013. I don't know how that happened. That was like, four months ago, I really have no idea how I let four months go by. Stuff was going on last fall, some crazy stuff went down in my life and I just, sort of, didn't do this show for a little while, but it turns out "a little while" was longer. So. I'm sorry for that. The show is happening. Here we are. I'm going to go back to doing my best to do this show every week. And I want to say thank you to everybody who's been pinging me, sending notes, sending Tweets, emails, asking me, "Hey, where's the show?" It really does mean a lot. And it lets me know that you guys really are out there and you are wanting to listen to the show. I think that, more than anything, really motivates me to just get it together, send out some emails, find some awesome guests, and record an episode.

So here we are, number 60. I do have two things that I want to ask of you, if you are a big fan of the show, if you like the show, there's two things I'd like to ask you to do. The first one is: Help me out, help us out in getting people to re-subscribe. I don't quite know how iTunes works now. I do know that two years ago, when I first started doing these podcasts, if a show stopped posting new episodes, then iTunes would stop trying to download them. Then if a new show got posted, no one would know it because iTunes didn't download it. iTunes is not the only way that people listen to the show, there's tons of other ways, including the 5by5.tv website and other ways. But a lot of people subscribe in iTunes. So you might have to... of course, if you're listening to this, then you've already somehow acquired the file. But you might have to go and tell your iTunes to re-subscribe. Or, really, what I'm asking is, "Hey, can you Tweet about the show or write a blog post, even, on your own blog? Or post to Facebook, to Google+, tell your friends, tell your colleagues at work. Get the word out about the show again. Because it may have changed, but it may be true that people are going to actually go and re-subscribe and that the audience will have dropped quite a bit because the software won't be downloading stuff. It will help a lot to get the show going again. To get the word out that there is a new episode.

The other thing I'd love to have to do is... the net Awards. The folks over in the UK who publish net magazine have an awards show and they haven't done a podcast award in awhile. They haven't done one since I started The Web Ahead. But they are this year. And, so, The Web Ahead was fabulously nominated for the category of Podcast of the Year, along with nine other shows. And I would love your vote. God, it feels weird to say that. I don't... I'm so torn about whether or not to do self-promotion. It's really sort of weird, annoying, when other people lobby for themselves, but that is how the world works. And I would love to have you go and vote in this category and tell the folks at the net Awards which podcasts you think are the most awesome. Perhaps The Web Ahead is the one that you like. I'll put a link in the show notes or you can go to thenetawards.com and vote in a whole bunch of different categories. Including, for people who've been on the show as guests, if you love their work, you can go vote for them in all sorts of other categories. But, enough, enough, enough of the business promotions. Let's do a show.

Today my guest is Doug Schepers. He is a web standards specialist over at the W3C. He works for the W3C. He's been very active in the SVG specification, making it happen and evangelizing it, explaining to people how to use it, and he works on a number of projects. One of which is webplatform.org where he and I have been working together on a lot of stuff. He's been talking a lot about this thing called annotations. I thought, it would be awesome to have him on the show to talk about annotations and also to really talk about the larger issue of, how do we have conversations on the web? Comments and all sorts of different ways that we can post of each other's websites and have a bigger conversation, a back-and-forth. Right now, that happens in certain spaces. Facebook, for example, very easily. There's a bigger issue of what's happening, how things used to happen, how things have evolved over time and then, what is it that we want? Those of us who... I mean, there's people who have imagined the web. We imagine the web and then we make the web into what we want it to be.

Hello, Doug, welcome to the show.

Doug

Hello Jen. Thanks for having me.

Jen

Let's talk about this. What is annotations and what is it that you've been thinking a lot about when it comes to conversations on the web?

Doug

I think it's a continuum. What is an annotation? To some degree, if I make a comment about a blog post, if I link to an article on Twitter and make a comment, or Facebook, or whatever, and I make a comment about it. I'm annotating that resource, I'm annotating that document, I'm annotating that blog post or whatever it is. Because I have a link to it and I have a comment about it. It's a slippery slop, "What is an annotation?" I did a little diagram based on what the different things are. A footnote is a kind of an annotation, right? It links into a specific part of a blog post. I'm sorry, links into a specific part of an article or whatever. It says something about that specific part. I think if you're going to break it down, to be a proper annotation and not just a comment, has to be talking about a specific part of a larger body of work. A footnote is like an annotation because it's pointing to a specific part of an article. But it's also usually published along with the article itself. In a book, an ebook, or on the web, or whatever. Footnotes are sort of canonical. Annotations are what somebody else is saying or what the author themselves are saying at a different time than the publication. It's another layer. Annotations are another layer of content, about the content, on top of the content, deeply linked into that content. It's that deep linking part that I think is the most interesting part, technically.

Jen

What have we been doing so far, do you think, on the web that resembles annotations?

Doug

Thank you for the leading question. [Both laugh] Comments are really a kind of annotation. Especially if you select some passage and you say, "Hey, you said this, and I have this to say about that other thing." But they're so disjointed. Commenting is one of the main things we do on the web. Besides reading. Our active contributions to the web are to share things and to comment on them. Most of what people do, when they aren't just reading, actually falls into the category of... not creating primary source documents, not creating blog posts, not creating articles or memes or whatever... it's commenting on something else. Or sharing something that somebody else has made. Even really sharing something is a comment on it. Especially if you tag something. Even just tagging something is a kind of annotation, right? You're making a meta comment about that thing. A lot of sites are turning comments off. I know Popular Science turned their comments off because the quality of the comments was so poor. It detracted from their content. Whereas, you've got other sites that are really embracing comments to an extraordinary extent. Like the Gawker properties, right? io9, or, you know, the whole thing. They've got this engine where... my wife reads the articles for the comments, not for the articles themselves, because they've really embraced how you can comment on things and how you can bubble up the best content. It actually becomes the content itself to some degree, right? It adds so much value to the original article that you've got this ecosystem of people out there who are creating content that happens to be referring to some other content. That, in itself, is most of the way to annotation. Then you've got awesome sites like Medium or Quartz or many others. New York Times has experimented with this. When a comment is at the bottom of a page, it's so abstracted out from the rest. They get off track, they start talking about other things that have nothing to do with the original article. If it's an even vaguely political topic, you've got the partisans jumping in, yelling at one another, how they're all idiots. You lose track with the content of the article. There's this viscerality, this immediacy, of actually commenting on something in its context. Imagine that you highlight something. Medium does this on a paragraph level. Rap Genius does this on their lyrics. They let you select something... or I think they way they do it, is they link to specific lines in a poem, in a song. And you click on it and you add an annotation about that. The best interface, as far as I'm concerned, is you select something and you get a little icon that pops up and you say, "Yeah, I want to comment on this," and it anchors it to that specific passage that you pointed at. When you have that, the quality of the comments becomes so much higher because you've got this immediacy. We've all been doing this for years. If you've ever used Word or Google Docs or anything like that, when you go through the peer review process, somebody says, "Oh, I think you should change this," or "Oh, I think weakens the argument," or whatever the comment they have. "Oh, this is a typo." They could change it themselves but it's actually, some people feel comfortable changing typos on a wiki or whatever, but they'll comment that somebody else has a typo or they'll comment that somebody else needs to re-word a passage. They'll suggest the change. A suggested change is an annotation, right? It's an interesting annotation. When you ask that somebody changes something, the thing that you're anchoring to... if I want to change the word "dog" to "fox", right? I've asked them to say, "Ok, well, take this word 'dog' and change it to 'fox'." What does that original anchor to? Does it still anchor to the word "dog", which has been changed in the original source? These get into the technical problems of an editing workflow. I actually feel, I'm getting really off track. [Laughs]

Jen

It's complicated. It feels like... I was thinking about this yesterday and I was thinking about... you know, it feels like comments, especially comments on a blog, a personal blog, were some of the original ways in which people... once blogging engines got started, like WordPress and Blogger before that, it works because the communities were small. There were maybe 40 people reading a post and six people decided they wanted to say something about it. People were using their real names, perhaps, so it felt like a conversation between seven people and that 40 people were watching. But it doesn't scale. I think it was probably YouTube, once you got to these giant sites, or, newspapers, something like The New York Times, where the last thing in the world, old school, newspaper-thinking-people, wanted was to open things up to comments. It's like, we don't need the riff raff saying all kinds of crazy things about our professional, well-thought-out reporting that took weeks to do. I remember railing against that and being like, "No! It's the future, it's the now!" Like, we do want to have a conversation. It's not that news comes from on high from this lofty world that's so special that no one should touch it. This should be a debate and a conversation that we're having. But then, like you said, things just devolved so quickly into crazy town.

Doug

Newspapers have always embraced comments, right? That's your letter to the editor. But they want to curate them. They want to cull the junk and they want to bring the cream of the crop to the attention of their readers. Somebody says, "Hey, I had this thought about this article that you published last week." And they publish that, right? Because of that lack of immediacy, you've got some temporal lag between the immediate reaction, you don't get this sort of spammy, troll-y, promotional thing. And they curated it, right? I think that's actually a key aspect of the whole commenting ecosystem. I'm going to say commenting because annotations and comments are slightly different but it's a similar thing. You really want to have this quality bubble up to the top. I think somehow Gawker manages to do that. Other sites have not. YouTube doesn't even try, right? It's somehow not quite enough to just... I think letting people vote up other people's posts is a useful mechanism. It's not perfect and you've also got this negative aspect of curation where the publisher decides what the voice is going to be. That's where I think annotations is super interesting, right? Let's say you're a political blogger, and I don't think you are. [Laughs] But let's say that you have this point of view that you want to get across. I come along and I tear apart your argument and you just decide, "You know what? I'm not going to let this well-thought-out comment go through because it weakens my argument. But I am going to let this crazy, oppositional point of view come through because it makes people on the other side of the issue look like idiots." And that's pernicious, right? That's insidious. You've got that going on. This notion that whoever controls the original source also controls the dialog and that's dangerous. This is why I like the idea of annotations. It's inherent in the idea of annotations, this indie web aspect of, "I want to control what I say. What channels it goes out to. I can't control who puts it into a different channel but I can control what channels I try to put it out into." I can actively publish in multiple channels. Say somebody has some article on global warming. I decide to comment on it. Actually, my comments are probably not that useful because I don't have really good information on it. But let's say a scientist wants to come along and say, "Actually, this has been disproven, you're just perpetuating a myth," blah blah blah. With an annotation, somebody reading that article might be following along with the article and say, "Oh yeah, I see where this anti-global-warming person is coming from. I really understand." If all that's there, if the only voice there, is the voice that the anti-global-warming person is saying, it's easy for people to get pulled in within a countervailing opinion or fact or whatever. They're reading along an article and they see this thing that seems like this compelling argument but then there's this highlight on it, right? When they mouse over it, they see that somebody has left a comment. They look at the comment. This comment from a well-respected scientist could come up, right? There is context. "Actually, this has been disproven. See this other link over here for a reference." That's the social aspect of annotations that I think is really powerful, right? Maybe that scientist has not published it directly on that site, right? Because that site for whatever reason doesn't want to have that voice there. He's published it on some third party service, right? Some annotation service that puts it on top of that site whether the site owner wants it or not. And here it comes into curation.

Jen

Talk more about that. Because, probably everybody listening knows how comments work, knows WordPress and YouTube and all that, and most of us, many of us, have checked out things like Quora, which I think is very interesting. Where they changed how commenting works, like you alluded to or were just talking about, where you can vote things up or you can respond directly to other people's comments. They end up, maybe, not in chronological order, they end up hopefully in some other kind of order. New York Times, for example, the editor's will pick certain comments and those comments are at the top and they're the only ones that are displayed by default. So when you start to read the comments, what you're reading is actually pretty interesting. But all of those systems are designed by the designers for that particular website. They're built by developers for those particular websites. I haven't seen a lot of open source software that does a much more complicated, and I don't know, whatever, mature way of commenting out there in the world that makes it easier for everybody to just do this. Perhaps that could happen. But like you just said, even if something like that happened, where WordPress had some amazing commenting system that was maybe Medium, or Medium meets Quora, or whatever. It still is controlled by the person who runs that particular website. What you're talking about, and what we've talked about previous in conversations that you and I have had, is a system that's really more of a new API or a new technology for the web that doesn't exist right now, that would make it possible for websites to have this layer of metadata and have this layer of connectivity between different sites. All this information shared between things like Twitter and Facebook as well as other websites, that it's not about one particular... it's not each site being it's own garden. It's a whole other thing. Talk a bit about what you've seen people are making, are building, right now, and how those work.

Doug

There's this nonprofit called Hypothesis and I'm in love with this nonprofit. These guys are great. Disclaimer, we're working with them to enable annotations, proper annotations, on webplatform.org and also on W3C specifications. I can get into that more later. I just wanted to make it clear that I know these guys. There are many other sites and many other projects. I don't have anything against them but the one I'm most familiar with is Hypothesis so I'm going to talk more about that. This was started by a guy named Dan Whaley and he said, and actually, I feel lame because of that global warming thing. That's actually where he started. That's why he started. Hypothesis, that global warming example. We want to be able to get this information out there to people. How can we have this curated information? So he started a nonprofit to do that, not a commercial company. They are actually contributing to this open source project called Annotator. Annotator started from the Open Knowledge Foundation. There's a guy named Nick Stenning who really started that and Dan Whaley and his team, Randall Leeds and a bunch of other guys, have been really re-architecting that to make it scale better on the web. There's e few problems that they come into, right? They want to be able to annotate any webpage and they do it in two ways. They have an annotation server on the backend and they have an annotation either extension of bookmarklet that you put into your browser and you select a passage and it triggers that to the annotation system. It opens up a little sidebar, you can type your comment, it sends it off and it publishes it on their site. But you don't have to publish on their site, right? They're open source. W3C is going to have its own engine, it's own annotation server. That's a key part of it, I think. Let's say that you are on Quora or Disqus or Livefyre or some individual site. Those are all publishers if you think about it, right? They're choosing what content they publish, either laxly or actively, they're choosing this. We have this notion nowadays that if you post something to Facebook, you're giving up that Facebook owns that. They're going to do whatever they're going to do with it. I like how Dropbox is expanding into the space of letting people store their data on Dropbox and not just have a folder on Dropbox but using it as as service. I think that what they're doing is really interesting. They claim that they're not going to be using people's data. The nice thing about an annotation service, why do I have to choose whether I'm going to put something on Twitter or Facebook or whatever. All that does is it builds up these silos. I don't mean anything against these companies. They provide a service, they're great. But they do silo-ize the content that gets put onto them. Why do I have to choose what my input to those... Twitter did this whole thing where they wouldn't let people post through particular apps, or you have to verify your app that you're going to post to. They wouldn't let certain services publish to Twitter from their competing service. TO me, that's really lame. [Laughs] I'm creating the content, right? I should be able to publish it, push it out, however I want, to whatever channel I want. Maybe I have a channel where I put things for my family and one for my coworkers and one for when I want to make just snarky, humorous comments and one where I'm making serious comments. Those don't have to be different services. I could make a serious comment and a really snarky, sarcastic comment that I'm hoping is funny, on the same article. I should be able to push out what I'm trying to do. I should be able to say who I'm pushing it out to. Conversely, actually, I should be able to publish my comment, my content, wherever I want. As easily as possible. An annotation system should let me choose, when I make the comment, which syndicators, which publishers this is going out to, which aggregators this is going out to. But then it should also be able to let the person who made the original content, the source content, also consume my content and maybe suck it into their commenting system, if they feel it adds value to their system. Just like you said, the curation that The New York Times is doing. They should know that I made a comment on their system, if the comment isn't something private that I'm only sharing with my coworkers or whatever. They should know that, "Hey, there's all this content that points back to us, that points back to some specific part of our content. We think there's value in that particular content. We'd like to publish that on our site." And they should be able to. It's this chain of attribution that I think is really important that you sometimes lose on sites like Pinterest or Tumblr or whatever. "Where did this thing originally come from?" To me, that's important. Who said this?" Give them credit or give them blame but point me back to the original source of this thing. That unique identifier for the original comment. On Twitter, if you click on the timestamp it takes you to the original post that you made about that thing. It's the permalink. That permalink doesn't have to be just one permalink, right? You should be able to say, "This is the original source of what has become syndicated, what has become shared, when it's passed around the web." I think that attribution is actually one of those things that's overlooked but is really important.

Jen

There's so much in everything you just said. There's so many things in there that we could just unwind. One of them, this idea that I could... I'm on Facebook, I say this that and the other, I'm on Twitter I say these other things, these other things just disappear. The idea that somehow, if instead, data could be, content could be floating around in the ether, in the internet, a bit more agnostic to where it got created originally. Cross-post to other places so that my own website immediately has everything I said on Facebook or everything I said on Twitter. Or maybe I don't want my Facebook stuff to be publish because it's more personal but I use my own website as a way to archive it. Or maybe I build a website for all my family and everybody who's my closest friends and family get to see everything. That it's the same so that I can have it 10 years from now, 20 years from now, even if Facebook disappears, or I decide to cancel my Facebook account, or Facebook turns into something I don't really want to be involved in anymore. It's an interesting concept. We're so used to imagining the world the way it is already and seeing it the way it is already. It's a bit confusing to understand it if things were totally different and a bit mind blowing. Like, "Wow. What if the content were just content and there was some sort of generalized API that we all got to use." Oh wait, I think we invented that! I think it's called HTML. [Both laugh] I think it's called HTML!

Doug

I think it's thing called the web. [Both laugh]

Jen

Right? Like, wait, how do we keep remembering what the web was invented for? The web was invented to take content from a wide diversity of sources that were completely, randomly, just from every which way. Different cultures, different languages, nobody's coordinating it, there's no central authority, everybody builds, marks up this content in this way that we agreed to mark it up and then it just works across the whole system. Anybody can read it, anybody can write it, anybody can share it. It doesn't go anywhere. It sticks around. That's the original vision of the web.

Doug

Tim Berners-Lee, his first browser was a read-write browser. It wasn't just a consuming browser, right? You didn't have bookmarks. You had your own homepage. I mean, Tim is the original indie web guy. [Laughs] Because he had this idea: You had your own homepage and you edited that, to add links to things you're interested in, and you organized it however you wanted, right? But you had links to the things that you wanted to store. That was your bookmarks, was your homepage. And anybody else could go to that, as well, and see the things that you were interested in. Later, browsers didn't really have that. There's bookmark sharing services but really the browser was all about you. About your consumption of the web and your history. Those are all useful concepts, but sharing bookmarks and tagging bookmarks, like you can in Firefox for example. These help re-contextualize. You have this thing where annotations was sort of Annotations goes back to Vannevar Bush, right? With his whole idea of the Memex. I don't want to get too far into it if people don't know about it. Basically the idea was, you have this big analog desk where you have micofiches of all the scientific papers that were being published and you made connections between two different publications and you left a little note about it. This was all stored and it was your external memory device. These annotations go back all the way to that, really. Of course, obviously, annotations go back much further. Not to get too historical on you, but the most annotated document in history is the Bible, right? In Jewish tradition, the Talmud is nothing but annotations on the Torah. I can't remember the name of it, but there's an equivalent on the Koran. The Analects of Confucius, which is the big collection of all of Confucius' teachings also has a scholarly tradition of people annotating the Analects of Confucius. They collected a giant, huge, multi-volume set of the best annotations on the Analects of Confucius. This was published a few years ago. It's in Chinese, so there's no chance I would be able to read it. But it's this long tradition of people commenting on something that they think is important and persistent. It's ironic in some ways and you allude to this. It's ironic that these writings which have come down to us from eons ago, right? Millennia ago, literally, sometimes. They're going to be around millennia around, and these comments we have in this modern era, where communication is something that we think we've really gotten right. We're so much more advanced than the ancients. But our stuff is going to be gone long before all these annotations on the Bible or the Torah or the Analects of Confucius. Long after our stuff is digital dust, those things are going to persist. That bothers me somehow. [Laughs] We should be able to have some sort of system where a real conversation can happen. And culture can be aggregated, culture can be accreted over time, like it should be, so that we actually can trace where this idea came from. We can refine things over time. We can improve our culture over time. It sounds kind of lofty, and maybe sort of abstract, I think that's what annotations can help us do. It can actually increase the growth of ideas and not the suppression of ideas. It can improve how we create our culture in a more conscious way. In a way that's more critical thinking.

I'm yammering on. On of the things about an annotation ecosystem. Can we talk about an annotation ecosystem for a second?

Jen

We started to talk about Hypothesis. I'm looking at, I'm right now on annotatorjs.org, the webpage. This is just to be practical for a moment about how things are at the moment, right?

Doug

Yeah, absolutely. Let's get back to some real world stuff.

Jen

This is an open source JavaScript library. I don't need to add this to The New York Times website. Which I have no commit access to. [Laughs] I don't need to add this to my own website, even, if I don't want to. I can add this to my browser and it gets installed into my browser itself. Then this JavaScript-y magic will let me, because I;m using a browser that has this thing installed, highlight any webpage on the entire internet and make a little comment and add tags and click a "Save" button. Then I'm saving it to their... they're running a server generously for us, is that true?

Doug

Right. That's correct.

Jen

Yeah, ok. They're running one server, although I could set up my own server or I could call my friend Doug and say, "Hey, did you get this server set up yet?" And I can use that server when I have permission to use it. So all this data gets saved on a server, on a web server someplace, but it's separate from the webpages and the websites that any person was actually, that I was actually making comments about or annotating.

Doug

That's exactly it, yeah.

Jen

So if I go to your website and I make a bunch of annotations, any time I go back to your website I could see those, I'm imagining. But if Brad goes to your website, he's not going to see my annotations, unless you... explain how this part works.

Doug

Let's say that you make an annotation on, let's say webplatform.org. If webplatform.org has its own annotation system, that's one endpoint, one publisher for that annotation. But it doesn't have to actually go there. You can say, "Send it to this other place as well." You can log in to multiple accounts at the the same and you can say, "I want this to actually be published on my blog." I'm going to say annotation server and I don't want people to get scared about that. It's nothing special. It's just a data store. An annotation server is just some place that provides a place to publish the comment, the annotation, and preferably provides an API to also retrieve that annotation. Let's say that you have an annotation server, which is really just a commenting system in your blog. Your blog engine might have a little annotation server that lets you go out and publish things through it. The same way you can have your own OpenID on your blog, you can have your own annotation server. It's already set up. A WordPress site is already set up to store comments. Why were blog posts... why can't they be your comment to your blog post or whatever, from elsewhere on the web. There's not reason why. You log in to whatever service you choose to use for your annotations and you've got this extension, right? And you log in to whatever service or services that you want to have running for your publication. You make a highlight. You make a comment on it. You choose whether it's a private annotation, something only for you, or a limited group of friends or colleagues or whatever, or if it's a public annotation. If you make a private annotation, you say, "Doug's comment here is idiotic." And you make a public annotation, "Doug's comment here is very interesting." And you make that public. Brad would see, "Doug's comment here is very interesting." [Laughs] And your friends would see, "Doug is an idiot."

Jen

Let's say I go and I annotate something on The New York Times website. The New York Times has never heard of annotations, they don't listen to the podcast.

Doug

They don't need to. They don't need to know anything about it. The other site doesn't need to know anything about it, right? In the same way that you could bookmark something without The New York Times knowing about it. Or let's use another site. Whatever. The New York Times.

Jen

It could be anybody. Let's say I go there and I make an annotation and I write up what I think. Then you go there, you have to go to the same article, and you are using the same JavaScript library, the same little tool, you've got it installed in your browser as well. Are you going to be able to see my annotation because your browsing with that extension and you're hooked up?

Doug

Yeah. That's a good point.

Jen

A public annotation. I mean, are we at the point yet?

Doug

I keep going off into these abstract notions about what it could be, what it should be, and blah blah blah. But yes, practically speaking, if you and I both have AnnotatorJS installed as an extension... I think they have a Chrome extension or you can do a bookmarklet. I don't think that it has an extension for other systems yet. The bookmarklet works just as well. You go there and you can choose to turn that on or off for any given site. There's a little speech bubble in your awesome bar. You click on that and it shows you if there are annotations on the webpage. Their idea is that you would be able to filter these somehow. It's not just an idea, you can filter comments, right? Let's say there's two comments. Ok, you can read both those comments, ok, that's fine. But then you can also filter it by user, by when the comment was made, by any number of other... who made the comment, when they made the comment, by keywords, by tags, whatever else. You can filter it. This all works today. I can go to New York Times and I can leave a public comment. Jen, you go there, you also have this extension, this AnnotatorJS extension, and you can see my comment if I made it public. And The New York Times doesn't have to know anything about this. In the same way that they don't know that I bookmarked their site, this is separate.

Jen

It reminds me a little bit of Daring Fireball. Not so much what Johnny Gruber did on his website, Daring Fireball. Because he has never had comments. He was like, "I don't want comments. There's no comments. I'm not doing comments. That's it. End of story." But he has so many readers and so many fans that a group of fans said to each other and themselves, they were like, "We want to be able to comment on John Gruber's blog." You can install, and I don't know the details and I don't know if it's even still working today, it might be, somebody could ping me on Twitter and I'll re-tweet it or something.You can add this bookmarklet/plugin/extension/whatever into your browser and then you've got Daring Fireball comments. When you g to daringfireball.net, you get to see all the comments that other people wrote, you get to make your own comment, you get to pretend like there's a commenting system on Daring Fireball even when there's not. Because it's, like, layered on the top of it, run separately. It's like that for annotations but it's for the whole web.

Doug

Exactly, exactly.

Jen

Not custom to one website, again, it's something that would work across the whole web.

Doug

AnnotatorJS works right now.

Jen

Yeah. And from a really practical, for people who are really practical, "What can I do today with this?" This actually sounds like a great tool to use on a web design development team. Because you could get your whole team to use it and then you could make a whole bunch of annotations and say, "This thing here, make this text bigger. I don't like the way this looks. This is the whole shade of blue. We need to move this over here. We need to re-write this paragraph." If you're on a team of people collaborating on something, that's way better than making screenshots and trying to put comments into some other system.

Doug

There's another interesting aspect. It doesn't just have to be text. There's another organization, down in Australia, that is also making an extension for Annotator. I love the name, it's called Annotorious. Which is just a fantastic name. [Laughs] Annotorious lets you annotate images. In the same way that Flickr... or many other sites now, but I think Flickr was very innovative here, if I recall correctly. It puts a little box around people's faces and you can identify who that is. That's an annotation, right? On SoundCloud, you can leave a comment at a particular time point on an audio stream. Most of the ones on SoundCloud are along the lines of, "Hey, cool, man." But you could have more interesting commentary on a timeline. Of course, you combine those those and you could have both the visual and the auditory. Something on a timeline, like a video, along different points in a video you could have annotations not only on that time point but on this particular part of the image at the same time.

Jen

For this podcast, I could have just put the link to the Daring Fireball thing, or someone else. Anybody that's listening to the show could go in and say, "Oh yeah, around minute 57, right here is where she was talking about this thing, here's the link." Bam.

Doug

Absolutely. Annotations are about different media. I was talking about more document things, but it's for different media. You could even have just an image of something. Let's say you have a webpage. You're designing a webpage. You just have an image that you have that you made, maybe in Photoshop or whatever. I think people still do that. Where they make a proof in Photoshop and then they try to make it into a webpage.

Jen

Right. [Both laugh] That's a different show. [Doug laughs]

Doug

You could annotate that image and say, "Ok, this needs to change, this needs to change." Or whatever. It's an awesome workflow tool, right? One of the things I use it for, right now, is... I'm one of these tab orders. I have way too many tabs open in my browser at any one time. I've tried to find ways of... it's sort of, I really do think it's a hoarding instinct, right? It's sort of an ADD thing. "Oh, there's all these things I'm interested in. I want to keep this around. I don't want to forget this." So what I've been doing is, I've been annotating. I've been skimming things and annotating things I want to come back to, or things that I thought were really useful, or things that I'd be inspired by for projects I'm working on. I've just been leaving quick comments on the website. These are just my personal annotations. I've just been saving them as annotations. And it's so much more satisfying than bookmarks. I bookmark something and never come back to it. Even if I tag it, it just never worked for me. But if I highlight a particular thing that interested me, and I can even aggregate all the things that I found interesting, I highlight it and I store it. This is great for research. If you're doing research on a topic, you can go look through pages and you can highlight those. You don't just need to bookmark them. You don't need to copy and paste that into some other file. It saves where you highlighted it from, it saves your comments about that, et cetera, your tags about it. You're doing a research paper or whatever you're doing, you're doing any kind of research, you're highlighting things. You're pulling them together. And at the end of it, you go into your own annotations and you have links to dozens, hundreds, of documents and pull quotes from all of them and your comments about those things and your tags about those things. And you filter them out. And you can do this all today using AnnotatorJS. I'm not trying to advocate for any particular solution because I think ultimately it's going to be an ecosystem of different annotation systems and I think AnnotatorJS is going to be one of those. Right now, you can filter out... I wanted to say, I found all the things on... all the annotations I made where I thought somebody did something cool that I'd like to do on webplatform.org. Like, I'd like to have something similar on webplatform.org. I could just go to my filter and I have tagged it "web platform". So I look for all my tags for annotations for "web platform" and I share those with other people. Like a tweet, each of them has a unique link, a unique address that I can send to somebody else. I can share something with somebody else. I can say, "This is the particular comment I made" and then they can follow the link. In that link there's a link back to the original document, so they can see it in context.

Jen

Right. There's probably many people listening right now who would say, "I use Evernote for that." Or there are some other things out there. Again, the difference I think, and I'll just reiterate it because I think this is a big point. Not just for this conversation but just in general, an understanding of where we are today, 2014, what it is that we're doing what our lives, how this culture's changing, what's important to us, what kind of future we want to build. Evernote's awesome. I use it. I used it today. I'm going to keep using it. But it's a service. And it's a service owned by a single corporation and who knows what's going to happen to them? Sometimes I'm going to pay them and I want to pay them because paying them means they're more likely to stick around. But also, what about if I can't afford to pay them? What if they sell themselves to somebody else and then they turn into something I really don't want to be involved with? What if I want to export their stuff? I think Evernote has a pretty great export, where you can actually export to HTML, which is awesome. But they could have been in a parallel universe where they didn't have that and I could get stuck in their proprietary system. It's not the web. The web itself, HTML is not owned by any one corporation. CERN gave their license to it up to the public domain. The web exists because we all get to use HTML, we all get to use CSS, we all get to use JavaScript. You can do all kinds of crazy stuff with it and nobody's going to stop you. I think that this annotation idea is something that needs to be lifted into the realm. Out of just simply a realm of a great service that a number of different vendors can provide you for however long they stick around, into a realm of, this is baked straight into the web. This is part of what the web is and this is part of how the web works and every website in the whole world uses it and every website, everywhere, there's no option about this. This is how it works.

Doug

It's not they use it, it's that we use it.

Jen

It's a feature in all the browsers and it's a feature in all the browsers that works the same way in every browser. It's not an extra feature in one browser, it's a feature that a;; the browsers have.

Doug

And which annotation service you use to subscribe to, or publish to, or whatever, that's up to the individual services, right? I think some of these features, like the robust anchoring. Where I say robust anchoring, that means being able to point to a particular point in a blog, even though it doesn't have an ID attached to it or a span attached to it or whatever, I can point to a particular passage in a blog. That's robust anchoring. Maybe it's in the multi-page version and maybe it's in the single page version. But it's still the same comment, applies to the same piece of content. I really want these things to be part of the browser. It's too hard for them to do it in script libraries now. I won't go into all the details about what problems they're running into with Annotator. Some sites follow events, so they can't get the event that they need to know about in order to leave the annotation. Sites that have certain kinds of dynamic content. There's other problems with that. I won't get into all that. Basically, this functionality should be part of browsers. And it was a part of a browser, right? This was a part of Netscape for about a week. [Both laugh] Netscape or Mosaic, I can't remember, but I'm pretty sure it was Netscape. Back around '91, '92, '93, I can't remember, they had this as a feature. They realized it was so popular of a feature, they didn't have the resources. These annotations needed to be stored somewhere and they didn't have the resources at that time to actually store these annotations. They realized it would be really an engineering challenge for them at that time. They turned it on, they asked for people to use it, to test it out, and then they turned it off, right? Because they didn't have the service. Marc Andreessen, one of the guys who actually prototyped this feature in Netscape back in the day, recently gave $15 million, perhaps as a penance, to Rap Genius. Because Rap Genius has this mission, and they also want to annotate everything. They're a funny company. But they really push forward this idea of annotation. Right now, they're annotating documents on their site. The bigger idea is to annotate things on any site. Even if it's just you want annotations on your own site, browsers could provide hooks, ways of doing these things that make it easier to do this, even to allow this functionality on your own site. It doesn't have to be a distributed system. It could even be something just for your website, you want to use these. I like the idea of annotation events.

Jen

You just said something that's like, a thing, right? I have built many websites. I've built many websites that have come and turned on, especially since I've built many websites using WordPress, using Drupal, using other systems, that's easy. I see something like Quora, I'm like, "That's much better." Or I see something like Medium. Or even A List Apart did some very interesting things when they redesigned last year with their comment system. And I think, "Oh my, I would love to do that instead, I like that better." But I don't have, you know, $25,000 to write a bunch of custom code for this website I'm working on, my own personal website, a client website. I would like to have something more complex for commenting. You think people should check out AnnotatorJS because you could just install this and set this up for a single website. Which would allow all users to go and annotate, mess around with it, without you having to build all this by yourself from scratch.

Doug

Yeah, you could just plug it into your website. That's what we're doing with webplatform.org and W3C.org. You could have it working for your site alone or you can have it working for whatever, right? You can install it into your browser or you can install it on your website. Either way it works.

Jen

Are there other projects people are working on other than AnnotatorJS? Or annotatorjs.org. This is not a brand new idea, right? This has been around for awhile. In some ways this reminds me of web mentions, which we were talking about on episode... when Jeremy Keith was on recently. Not recently, a bazillion years ago, in September. Episode 56. We were talking about the indie web movement and this waking up to, "Oh yeah, let's make sure we don't lose this web where many, many, many, many, many people know how to build a website and know how to own their own data and this is considered important." And web mentions being what Jeremy's enabled on his own website where he has not wanted to have comments. He's not had comments on his website forever. But you can use, and I'll out all this stuff into the show notes, you can use this certain technique and change a little bit of the markup... Jeremy changed a little bit of the markup on his website, and he put a little form on there, so people can basically go to their website. If I want to leave a comment, quote unquote, on Jeremy's blog, or I want to say something about Jeremy's blog post that he's written, I can go to my website and write up a blog post, or write a comment, whatever I want, and the two websites communicate with each other and my comment ends up on his blog post. Kind of automatically, where he's publishing my content on his site as well as the original is on my site. That's a way to encourage people to have a conversation, have commenting but not where my comment goes and lives in Jeremy's CMS, but my comment lives on my website. Pingback, you know, it's related to pingback. These things are all connected to each other, conceptually. Technically they're different attempts to solve this problem that's not yet been solved.

Doug

That's the whole thing I was talking about with federation and aggregation and syndication. It touches on all of that. Web mention is a cool idea, like you said, none of this has really been technically tackled yet. Annotation is the larger topic that i'm interested in, but there are all sorts of little things that you would want to standardize in order to enable annotation. The idea of something like web mention is one of those. As a plug for what we're doing, W3C is actually holding a workshop on this next month. If you look for "W3C annotation workshop" on your favorite search engine it should pull up our workshop. It's happening, actually, the day before the IAnnotate workshop, which is on annotations, which is in San Francisco. Our thing is April 2nd. At our thing, we're going to be talking about, "What should a W3C working group do in order to enable annotations?" I talked about robust anchoring and the idea of having a data model is super important so that no matter what my annotation is, no matter where it comes from, somebody could consume it and pull it into their own blog and have it work with the kind of formatting they need for their blog or whatever. And it could identify annotations and pull them in. Footnotes is another idea. I mean, annotations and footnotes, the historicity is different but a lot of the technical challenges are the same. This should enable footnotes. Like you said, there should be an API so that people can retrieve annotations from elsewhere and pull them into their own content. There's a whole bunch of little technical bits that need to be solved. Some of them will help us just for annotation and some of them will help us for other things as well, and web mention is one of those things that, it's part of the problem. Annotations is the larger problem.

Jen

It seems like if this whole thing is thought through from the perspective of several different use cases and a standard is written that can cover those use cases and provide a technology foundation. Like, "This is the way we do this." Then that standard could be used by things like Evernote or Goodreads, which is another... Goodreads, this service that lets you annotate a book and share those annotations with other people who are also reading that book. Rather than having a whole bunch of different services that are doing something similar but also very different... have completely different, custom-made technologies or APIs that their own developers wrote from scratch every time, solving the same problem over and over again in different ways. Where is that data going to be in 100 years, or in 500 years? It makes no sense. No one's going to be able to read any of this stuff. It's all going to be gone. But if there's a standard and then Goodreads can use the standard, Evernote can use the standard, and everybody can get on board with that same standard.

Doug

And it decreases the amount of work that they have to do. These are not trivial problems. Robust anchoring is not a trivial problem.

Jen

Right. [Laughs]

Doug

If you have this system across browsers... I say browsers, but it could be your ereader software, right? Your ebook reader software for your digital books, or epubs, or Kindle bookstore, whatever.

Jen

A car interface. The voice interface in your car that's going to process webpages. I mean, a browser is a lot of different things. Your browser on your cheap mobile phone is different.

Doug

In your TV. Yeah, et cetera. Your mobile device, all these other things. But if they all have the same underlying functionality of being able to do robust anchoring. First off, it's going to be a lot faster than a JavaScript implementation. JavaScript's fast, but it's never going to be faster than native code. There's all sorts of other optimizations that browsers could do for that. But having it in the browser, in the standard, means that everybody has an incentive to do it roughly the same way and they all save engineering costs by doing it that way. And everyone could just use the system. Medium could just use the system. It makes their own, it makes the work of web developers so much easier, to just reuse the bits that are in the browser. Imagine if you had to, I don't know...

Jen

... Invent HTTP every time you wanted to build a website?

Doug

Or even just, every time you wanted to make a website, you had to invent a way to add a hyperlink, right? [Laughs] Every time.

Jen

Yeah, and every website around the world used a different way to link to other websites.

Doug

Then you couldn't have screen readers that go through and find all the links. This is what standards are for. Is to enable services to build on top of the lowest common denominator of functionality. If Evernote and Twitter and my Kindle were all able to use this service... not service, I'm sorry, use this functionality... they could each provide their own service built on top of the functionality. Now, getting browsers to actually implement these things, that's another matter. Can we get browsers to realize that... actually this is one of those fundamental things that should be there, that should be there on the web, not just on the web, but should be a part of the web. This interconnection. This idea of commenting is so integral, in my opinion, is so fundamentally part of the web experience, it should be a part of browsers. Standards is funny, right? Because it's a tradeoff of getting everything you want or getting nothing that you want. [Both laugh] We have to find those things that really are going to be handled best by browsers and try to make a set of specs that the browsers can implement to make it incrementally easier to do these things, which at the same time allowing for innovation in other aspects. Each one of them right have a different UI for activating an annotation. But all of them should provide some way of doing that, right? Is it going to be in a sidebar, or is it going to be in little sticky notes, or whatever.

Jen

Right. That could be designed on a case-by-case basis. But the data model underneath it all.

Doug

The data model and the annotation.

Jen

The protocol.

Doug

Well, I wouldn't say protocol, honestly.

Jen

No?

Doug

No, because a protocol is REST, right? A protocol is HTTP. You're posting the same way you'd post anything else, right? But what an annotation looks like, how you say, "This part of the annotation is the quote that I'm pulling from the article." This is the blockquote, effectively. This part is my contribution. These are the tags. This is the link to the original article.

Jen

Here are the privacy settings I want to put on this.

Doug

Exactly.

Jen

Because we have blockquotes, that's not so hard to imagine. But privacy settings in a standard? That would be big.

Doug

Well, there's all sorts of privacy stuff going on in standards and security stuff. But being able to say... privacy is a big part of it. Because what I'm doing, I'm looking at your website because I'm a lawyer working on behalf of a client who thinks that you stole some of their content. I'm trying to point out points of similarity with their content and I want to highlight these things and I want to make comments on them. I don't want the target website to know that I'm doing this, right? I want to share this with my clients but I don't want to share it with that website. Or maybe I'm a dissident and I want to make some comment on some government web site and I don't want them to know that I even made that comment. Not just what the comment was or who made it but that I even made a comment, right? Privacy's a serious issue here. Maybe it's just because I'm a private person and I want to keep my own personal notes. That's also a legitimate thing. I should be able to share or not share with whom I want to share and keep from everybody else what I have to say.

Jen

Yeah. So, what is this going to take? There's a working group charter. Explain for people, especially for people who don't know really how the W3C works or how standards get started at the very beginning of the birth of an idea. What's the process and where is annotations at?

Doug

Yeah, I don't want to get too complicated about that. [Both laugh] Standards come around when a number of people have recognized a problem and they agree that they're going to solve it in the same way. That's ultimately what happens. Sometimes a company comes up with a cool idea and they show it to other people and the other people say, "Yeah, that's a cool idea, we'll do that in our browser, too." One browser says, "We'll do web animations." And the other browser companies say, "Yeah, we'll do those, too. Because we think there's value in that." In this case, it actually started, the standardization part started through something called the W3C Community Group. A Community Group is a group that anybody, for free, can join, and it's a forum to discuss a topic and maybe write a specification. A specification details how a browser would implement something, You might write a specification about it. At some point when you feel that your work is mature enough, you come to the W3C and say, "We'd like to see if you guys are interested in making this an actual standard." That's one way that I think is actually egalitarian. I like the idea of community groups. Let's let the best ideas flourish there and foster, give them a place for people to talk about it. Then bring it to W3C. The other way is, like I said, a browser vendor comes up with something and they implement it and they ship it, then they make a standard, and other browsers implement that as well. There are other ways. It could be that several companies come together and say, "We have this problem." They come to the W3C directly. But anyway. This one got started because some academics... I say academic in a nice way, I don't mean it in a nasty way to say "academics". But these people were working in biomedical engineering or working in physics or whatever. Were working in historical documents. They said, "We want to be able to exchange our annotations. All of us have a different system." All had different software. Not necessarily browser stuff, but all had different software, and they wanted to be able to exchange their annotations between those different pieces of software and publishing systems. They came to W3C and they worked up this RDF model. It's the Open Annotations Community Group. They worked up this model that, when I first looked at it, I thought, "That's too complex." But then, when I looked at it further, I was like, "It's modular, so you can actually pick and chose which pieces you would publish for any given annotation." I still think it should be simpler. But I think that their model, their underlying model is good. I like the semantic web. It's fine. It's sometimes a different community. I'm really more of a browser guy. The stuff I do is in the realm of JavaScript, HTML, SVG, CSS, and APIs, right? I approach the problem from that perspective. I looked at what they were doing and I looked at what Hypothesis had done and I said, "There's a way of..." And Hypothesis had been a part of that community group as well. And I said, "There's something here that can be standardized." So I talked with some other folks and I said, "Hey, let's start putting together a charter." A charter is basically a description of what a group is gonna work on. It's a document that say, "A particular working group is going to work on these particular deliverables." We actually have to do that because when a company decides to join a working group... things happen in working groups. Particular specs get develops by a particular working group. And that working group has members. And those members are either W3C members or invited experts or whomever. And they're making a patent commitment when they join that working group. That's actually a really key thing. The web is free. The web is open. The way W3C keeps the web open is we have a royalty-free patent policy. That means that if you are in a working group and you contribute something, you're not just putting that in here so that if somebody implements it, you get royalties. You get paid for your patent contribution. This is how some other standards organization does it. This is how MPEG works, right? No, with W3C you're joining this group and you realize, "Hey, by all working together and by all pooling all of our patents together, and letting anybody implement our stuff, the network effects are tremendous." And that's what makes the web better. We all benefit by that, right? By all pooling our intellectual property together and sharing it with the world. That's W3C's royalty-free patent policy. Go ahead, sorry.

Jen

So when a person joins a W3C working group, they sign a document that says, "I'm giving up my rights to any patent claims and putting them into the public domain"?

Doug

Not exactly. It's an implicit thing. They don't have to sign anything. But the more important part is that they're not giving up their patent claim. Their patent claim still exists. Let's say I had a feature on annotations, right? I had a patent on some robust anchoring thing. For anyone else who's using robust anchoring in a way that's not defined by the specification, my patent claim still applies. I can still either sue them or ask them for royalties or whatever. But if they're using it in a way that's defined by the specification, then they have a royalty-free grant for that patent. This is a really subtle point. But basically it means that our specifications are strongly royalty-free, right? That's a key part of it. The character defines... this is getting into silly details... but the charter defines what the working group is going to work on. Because no company wants to join something and just have the idea that any of their patents are subject to being royalty-free. They're selective in what they do. This is a challenge in standards but it's how we get everybody to work together. In this case, we're making a workshop where we're going to gather together the ideas. We make the character. We say, "These are the problems that this working group needs to solve." Another useful thing about charters is not just to define the legal characteristics but also it's a way to focus the group. "Hey, there's a ton of things we could do for annotations." But what are the most important ones? What do we really need to solve first? We're going to focus the group. The character is really the group's focus. From a legal perspective and also just from a "getting stuff done" perspective. We're going to put together a charter and then we ask out members, "Should we or should we not start this working group?" Sometimes they say "no", usually they say "yes". So we start a working group. Then companies join and we all work together on a specification. In this case, I think there's probably going to be about 3 or 4 different specifications that would come out of the annotations working group. You can see what those different specifications would be in the charter. Everybody gets together, we have discussions, we have a lot of email discussion, we usually have tele-cons once a week or so where we discuss particular issues, we have editors for specifications. In this case, we're also going to be... like I said, we going to have an experimental annotation interface, we're going to actually going to use this annotation interface to let people comment on the spec directly in the spec. Which will be a really cool way of doing it so that people don't have to join the mailing list, they can actually leave annotations for their comments. We hope that feedback mechanism should make it easier for people to work with our specs. All along, people are making experimental implementations, right? But once we think that, "Yeah, this spec is done," we put it into something called "last call". Anybody's welcome to comment throughout the process of creating these documents but last call is when we say, "We think we're done. What do you guys think? Do you think we're done?" In last call you typically get lots of comments that frustrate you, and you have to go back to the drawing board. That is the collaborative nature of standards, right? You want to get the best comments from around the world to make the best possible specification. That's why we're open and transparent. The developers are the ones who are going to have to live with it, so it's really crucial that they actually get involved. Anyway, we go to last call. At that point, that's when browsers... after we get out of last call, we go into something called "candidate recommendation". All this may be changing soon, actually, by the way, but I'm not going to get into that. The idea is, once we think it's stable, that's when browsers really considering shipping this, shipping a feature. It's not always browsers, but in the world I deal with, it's usually browsers. That's when they consider shipping a feature. We make test cases about it. You know, like Test the Web Forward, sort of test-crowd-sourcing, which is a really awesome thing that Adobe started and that W3C is running now. The idea is you want to have as many tests as possible so that the implementation's all do exactly the same things so that developers and designers don't run into those frustrating differences between browsers. Which everyone hates, right? Which increases your time for development many fold.

Jen

Everybody's gotta do it the same way.

Doug

Yes. And that's what the test suites are for. Once we satisfy ourselves that at least two implementations for every test have been passed, for every feature, have been passed. In other words, we have two different browsers that pass each test. It doesn't have to be the same two browsers but that's the goal. To have all the browsers passing all the tests. When we feel that it's sufficiently stable and sufficiently interoperable, that's when we make it a recommendation. What we call our "standards", our recommendations. At that point, all the patent stuff kicks in. When it becomes a recommendation, that's when all the patent commitments are made. That's basically how it's going to work. We're going to have this workshop, we're going to define the charter, we're going to try to get all the companies that are doing annotations to contribute to our working group, tell us how it should be done, help us actually get it done, and then we make a standard on it. I've seen ideas go from working group to in the browsers in six to eight months. Usually it takes longer. I'm hoping this is one... because the stuff we'd be defining wouldn't be too complicated to implements... I'm hoping we would have this done in a year, so we could actually, a year from now, or a year from when the working group starts, we could actually have this functionality in browsers. Most likely it's going to be more like a year to two years before you see it. If we're successful. A couple years before you're going to see it in every browser, every modern browser. That's kind of how it works.

Jen

That would be amazing.

Doug

It would be fantastic. I think it would be revolutionary. I think that if we did this, you would see annotation services pop up in the same... it would be a growth period for annotations. I think that once the functionality to do this easily became real, then you'd have all these services that are offering different things. Maybe you'd have one for climate scientists and one for chemical engineers and one for lawyers. Not one, five for lawyers, 10 for scientists, or whatever, right? You'd have these services that do it. You'd have 100 for students. Different services that they want to use. All those would be competing and the best ones... you'd get the one or two best services out of all those services that launched because of this. I really want this to happen. [Laughs] I really want to see this growth of annotations because we've made a standard that makes it easier for them to do that.

Jen

Yeah. From my perspective it seems much more realistic that a standard would get written and then the web itself would change and 20 years from now we'd look back and be like, "That's right, the web didn't used to do this thing that we actually think is so core and central to our experience on the web in 2030." Like, "Wasn't it so weird before we had this thing?" That's much more realistic than, say, one open-source community is able to build a JavaScript library that's so awesome that it just becomes really, really popular. That could happen, we see that happen with things like... name a JavaScript library, jQuery or something. There's just something about this that seems to me to be so integral to the nature of the web itself that it makes a lot of sense to me when there's talk of it becoming a standard itself. Becoming just baked right into the web.

Doug

Honestly, I think that's what it's going to take. I could be wrong about this, right? But I think that... you have services like Twitter, right? That sort of changed how people were doing a lot of things. That happens. W3C didn't do anything to make Twitter happen. At first, Twitter wasn't doing anything that was particularly interesting, web-wise. They just provided a service. They didn't even know what they were doing when they did it. They thought they were doing something else. [Laughs]

Jen

Most of us who joined Twitter were like, "This is stupid. I don't want to join this. I don't know what this is for." And then you joined it out of peer pressure and then you were like, "Oh. This completely changes my entire relationship with the web." Right? [Laughs]

Doug

Twitter didn't come up with the idea of retweets. I don't even think they came up with @ names.

Jen

@ replies, yeah. Yeah, no, that happened naturally.

Doug

They didn't come up with hashtags, right? Other people came up with these ideas. They started using these terms on their service. And they integrated those into their service. I think that's what's going to happen with annotations. You're going to start with a few core things that you can do with annotations. I think we're going to be stunned by what people will do with them in the future. Annotating song lyrics and things like that, like Rap Genius is doing, that's really awesome. Annotating documents for research, linking two documents together, having an annotation that's actually... with one anchor in one document and one anchor in another document... all these things are cool but it's not... the devil's in the details, right? There's nothing special about Twitter, right? Twitter is just a short message that you can send and can have links. That's it, right? Then it grew into this phenomenon. I think there's going to be services like that, that emerge for annotations. I want to say something about the ecosystem. I think this is the right time to say something about the ecosystem. In order to have an ecosystem, we need to have the basic functionality. It needs to be... it doesn't need to be in browsers, but it's best if it's in browsers. We can do it with JavaScript but kind of clumsily. I will say that you need to have the basic features in browsers then you also have to have a set of services that let people take advantage of those features. Like I said, for doctors and lawyers and such. For students, for whomever, rights? For whatever niche. For comedians. Whatever niche that you want to...

Jen

That's something that probably Facebook would add. Like, Facebook would add this.

Doug

Absolutely. I imagine that Facebook and Twitter will add annotations to their systems. To what they offer. If I were them, that's what I would do. When I say browsers, the functionality is not just in browsers but also, again, e-book readers and all these other things. There's the functionality in the reading system, there's the service that lets you distribute it, or aggregate it, or whatever you're going to do. The third part of the ecosystem is the people actually making the content, the people actually making the annotations. You say annotations and it almost sounds academic and abstract. But, first off, people have been commenting for however long we've had commenting systems on the web. What, 15-20 years? We've already got a set of people who know about that. But then we've also got an interesting set of people that have actually... are really familiar with the details of annotation, with highlighting, with tagging, and, like, doing critical thinking about what they're reading. Any student, or most students, who are in English classes today, you've got this whole generation of people who are doing this thing called "close reading". Teachers are teaching them how to specifically do annotations in books, right? Physical books. Highlighting things. Asking questions about that particular content. Like, tagging it with keywords. All sorts of categorizing it. Basically doing this kind of critical thinking. I think we're raising a generation of annotators and once we let them do that on the web, they're going to apply those skills that they learned in the English classes to other disciplines and just to even, just having fun on the web, and making jokes about song lyrics or mocking somebody else or whatever. But the point is that these people... we've already got this generation of people who know how to do annotations. We just aren't giving them the tools to do that on the web yet.

Jen

Yeah.

Doug

I think that those three parts of the ecosystem are the critical things. I think we've got everything in place, we just need to agree about how we're doing it.

Jen

Let me ask you about transclusion and Ted Nelson and... it's weird. I just, the last couple of years I feel like I spend more and more time thinking about and studying and researching the history of the web and the ideas that were around before the web even was around. Hypermedia, hypertexts, and it just feels like we still haven't completely implemented those ideas. One of them is this idea that Ted Nelson had. Ted Nelson being the guy who coined the term "hypertext", "hyperlink", "hypermedia", and back in the 60s...? I should have Wikipedia open.

Doug

Yeah, in the 60s. Last 60s.

Jen

He had this idea of transclusion. I know he, in some ways, he's frustrated and a bit... angry? Bitter? About the web. Because the web isn't what he imagined. That the web is a whole bunch of separate documents that are linked to each other but it doesn't have this element of conversation in the way that he imagined. His idea around transclusion is that... if I'm on the website and I say something awesome, and then you, Doug, want to talk about that. You want to write about what I've said, and you want to quote what I said. Right now you can copy and paste, right? You go to my blog, you copy, you paste, you mark it up with a blockquote. But those two things, those two documents, are fairly separate. We have these ideas of linking them together. We've talked about some of those ideas already in the show. But really, they're separate. When my website, my domain name expires, or my server blows up, it disappears. But his idea of transclusion is more like... you wouldn't, in your blog post, copy paste my quote. You would actually transclude it in. You would actually point to the page, to my page, and my page would end up over there on your page. Iframes are perhaps the closest idea. But it's not quite iFrames, either. It's something about... that my content actually becomes part of your content. Then if I change what I wrote, if I edit it and make an update or something, the new version is the one that ends up in your viewport, the viewport of looking at your website. You didn't make an old, static copy. You actually just included the living, breathing thing itself. In some ways, annotations reminds me a bit of that idea. That there's this other layer of conversation that isn't about separate silos, in separate web pages, in separate web documents. It's about an ecosystem of commenting and sharing and communicating that is separate from any one service, or separate from any one domain name. It becomes... I don't know. Now I'm being very abstract. I don't know if you have any thoughts about transclusion.

Doug

I like transclusion. I like his ideas. I don't want to disrespect him because I think he's brilliant. But I also have to say that... it's sort of like, when you come up with an idea, you're contributing it into culture. Then culture does with that idea what they want to do. This reminds me a little bit of an author who's upset that the movie version of their book isn't exactly the same as their book. But it couldn't be. Because a movie is different than a book. Even just down to... when I'm reading a book, I imagine what the characters look like. And in the movie they look different. They're all different from what the author thought of them as looking like. When you put something into culture, it transforms. He put his idea into culture and it transformed into something that was a workable model. I think that he is absolutely the person who really brought forward the idea of hypertext and transclusion but there's some aspects of that idea wouldn't work well. I don't like the idea that if I wrote something... ok, you wrote something and then I quote it and I start talking about it and I transclude your quote. And then you change it. Does that mean that I have to update my document? You did say what you said at one moment, right?

Jen

Yeah.

Doug

That, to me, that's a fundamentally... that is out there. You put that out there. I don't like the revisionist idea that somebody can change something and de-contextualize everything that originally contextualized it. Should there be another version? Memento is actually a really interesting protocol, talking about protocols. There's the Memento protocol, which has just become and RFC. I don't remember which RFC number. But, at the IETF. The idea is that you can point to something as it was written at a particular point in time. When I use archive.org and the Wayback Machine to look at what a website said at a particular time, right? Things can change on the web. The technical infrastructure to have it otherwise, to have it centralized, which was his idea. His idea was of a centralized web. Where links didn't break because somebody maintained those links. That's a much more complicated, technical enterprise than lettering everybody do their own thing. In some ways, what Tim did, Tim Berners-Lee, is both radical and brilliant. He had the idea of one-way links. Other hypertext systems at the time, in the late 80s when he was doing this, other hypertext systems had a centralized repository for the links and that's where the links were maintained. It's not very egalitarian, right? That's a curated web. That's sort of what Ted Nelson had in mind, was a curated web. From my interpretation. He can feel free to send me a nasty letter, telling me how wrong I am. [Laughs] But it's this idea of curation, right? Where a link is between two different things and to come degree both parties have to agree that link will exist and persist. With that link, that's how transclusion is possible. That link is not just a link on your... Tim made links that, by their very nature, because they were only on one document, they were only outbound links, they could break. The web we have today,for better or for worse, wouldn't have emerged if we had some centralized index of all the links. In the same way that we have a centralized index of all the domain names, right? Fundamentally I just don't see how the web could have emerged. Maybe it would have been a better web if it was curated but somehow I don't think so. Somehow I think that the egalitarian, radical, ugly, messy web that we have today is a web that more accurately reflects society than this curated web. Another aspect of his idea of transclusion is not just that I would be able to link to some part of your site and bring it into my site. It's that I would pay you for that content. There's an idea of micro payments inherent in that it's the idea of publishing, right? I'm republishing something that you wrote. When I'm transcluding it, I'm giving you some fraction of my income. I love the idea of micro payments. There's something called PaySworm which Manu Sporny is pushing forward, and there could be other kinds of web payments. W3C is actually having a workshop on web payments coming up. The idea of transcluded content... there's sort of a... not DRM-y, DRM exactly, but it's sort of DRM-esque, idea of the person who publishes the content controls the content and they control who can transclude their content. First off, technically it's a difficult problem, and second off I'm not sure if that leads to a better web. I actually prefer a web that's more distributed. It's not that his ideas haven't been realized. Some of them haven't been realized. I think in some ways, more that his idea didn't really fit the marketplace and the culture in which it emerged, to some degree. Some of his ideas I don't think are what we want to use. I love the idea of transclusion but how frustrating it must be if somebody changes something after you've already quoted it? I don't know. I guess that while I really appreciate some of his ideas, I think they were the seed of the idea and I think that what has grown from the seed is different by necessity.

Jen

Yeah, well, and I think all of this stuff is much bigger than any of us, any one of us, and many of what has happened over the last 50 years, much of what's happened is almost inevitable. There's this argument that you can debate and gets written about quite a lot. Is it that the technology got to a certain place and so therefore all of our culture has shifted? Lots of times people will hate technology because of that. They'll look at the mobile phone and they'll be like, "I remember how life was before the mobile phone and I don't like how it is now." It voices the idea that, culture is changing, and humanity is changing, the way human beings live our lives is changing and those changes are sort of unstoppable and the technology is being created out of a demand for... like, we're going down this road, somebody's going to invent this phone because this road demands that this phone is going to exist. It's a big of a philosophical conundrum of which way it is. It's probably both.

Doug

It's both.

Jen

Yeah. But there's something in, everything from the 40s, 50s, 60s, 70s, 80s, 90s, about inventing this web that we now work on and are continuing to evolve that... a lot of those ideas have been implemented, a lot of those ideas have not yet been implemented. That's what makes me fascinated, is like, "Ok, if we go back through all the ideas that have not yet been implemented... is there anything good in there that we should snatch up and do now?"

Doug

Absolutely.

Jen

Now that the computers are more powerful, now that we've already got the first part already done. Now that everybody knows how to use a website. What is it that could be more complicated? It feels like Twitter, in a way, came out of this need and demand for another layer of conversation. I think a lot of us had blogs that were, sort of, the nature of the nature of the content of the blog was, "Hey! It's Friday. Here's this cool thing I found. Oh, look, here's another cool thing I found. Here's a cool thing I found." Because how else did we find cool things and we found them by reading each other's websites with the cool things on them. Daring Fireball is an example of a website like that, that's sort of stuck around. But there used to be many, many, many of them and many, many, many of us have stopped making websites like that because there's no point. Instead, we use Twitter. Instead, we follow... and because it's so quick. It's just like, you just click the link and you copy and whole thing and you open Twitter and you make a new tweet and you type one sentence, you paste the link and you hit save, and you're done. You don't have to open up your blogging engine and go over to your website and log in and write something.

Doug

And think about something. [Laughs]

Jen

Right, yeah. [Both laugh] And it's dangerous. And you could make arguments the quality of the content is low because it's too easy and we end up saying snarky things that later we regret and you end up in flame wars that you didn't mean to get into. There's pros and cons. But there's something... it's just so easy and fluid to just quickly make a comment about something. All of that activity went from the harder-to-build blog or website. "I'm gonna open up my coding program and hand-code an HTML page to tell you about this other website that I like and then I'm going to FTP that webpage to my server." [Laughs]

Doug

Right.

Jen

We don't do that anymore. You just, boom, tweet, boom, done. I'm standing in line for the grocery store and I tweeted about the cookies that are on the shelf. Done. There is something, I feel like it's still... Twitter unleashed a whole world of stuff that's not really... Twitter's awesome. You could argue that Twitter's not awesome instead. But there's something there. There's the seed of something there that just, it feels like we want that so badly, we need that so badly. We need that to not be a single company that's dependent on VC funding and advertising revenue. [Doug laughs] We need that. A bigger conversation.

Doug

Right.

Jen

And we need that to be a conversation that's easy enough to use that people who have very little technical skill and have no desire to learn any more technical skill can use it. Because instead they don't even know how to use Twitter. Right now, they're all, they just, they use Facebook, or they do nothing. They don't participate at all. How could they be participating in a way that's... there you are.

Doug

I think annotations is pretty easy to use. The idea of selecting something and then leaving a comment about it. Without even really... I think that's pretty easy. Older people that aren't techno-savvy aside, I'm not sure, actually, personally, I want to hear the comments from somebody who doesn't know how to point and click. [Laughs] Or not even point and click, right? Like, you're on your tablet and you just drag your finger across something and then you either type or speak something. There's speech recognition APIs now. You can just say what you wanted to say. Or add a picture about it. Or whatever, right? You pointed out an interesting thing. How culture has shifted. We might lament how there was a, what? Ten, twelve year golden era of where people were actually, had blogs and were actually publishing on their blogs and Twitter has ruined all that. And gosh, when I was a kid. And all of us are just using Twitter and we're not posting to our blogs. It's true. It took me forever to get around to writing a blog post. I just blogged today, because I was going on your show, I blogged about my ideas about annotations. On scheppers.cc. If you look for "Doug Schepers" or "shepazu" and, you know, "annotations" and "W3C" you'll probably stumble upon it. And then you can add a link...

Jen

Yeah, I'm putting a link in the show notes. By the way, the show notes, also known as "the annotations for this episode" [both laugh] can be found in a big, long old school list at 5by5.tv/webahead/60, the number 6-0 because this is episode 60. I've been cramming a good zillion links in there to all kinds of things we've been talking about. Including this blog post that you just wrote.

Doug

So this blog post, right? It took it out of me, this thing. It used to be I could knock out a blog post and now, I really had to sit down and focus and do a blog post. You could argue that we, maybe some critical thinking that might not be... but then, maybe it's different. Maybe the amount of time that... maybe a tweet to a particular, with an insight about a particular new service or whatever, actually saves the reader a lot of time. It actually makes it more likely that we're going to comment. Like you said, it decreases the barrier, the overhead. I think that there's room for both long form content and annotations and tweets and who's to say that an annotation can't be long form? Have you ever read Nicholson Baker's The Mezzanine? The Mezzanine, the whole action that takes place during the book is... this guy is going up an escalator from the mezzanine, from the ground floor. Everything else in the book, this 200 page book, whatever, 150 page book... is just footnotes, which are annotations, effectively, on what he's thinking about while he's going up. The whole book is itself... so, certainly, long annotations could be long form. When we're lamenting the golden era of the long form blog as opposed to micro blogs, like Twitter, or maybe annotations. You know, the novel, how long has the novel been around? A few hundred years? And we think of this as something that's sacrosanct. But the novel, while the novel isn't going away, a lot fewer novels are being published now than were being published a few years ago, and certainly print published. Print publishing, I wouldn't say it's in trouble, but it's not what it used to be, right? These things change. They change according to culture, they change according to fads. They change. That is part of culture. I think that annotations actually will help us in both our long form and in our short form communications. I think that they will help us be more critical about long form content because we can actually pick it apart and analyze it and actually have meaningful things to say about it that can change the content itself sometimes. I think that being able to share things super easily will help the distribution aspect of that content. And just getting your thoughts out there. I think that annotations will serve both of those purposes. Both the ease of distribution and transmission and the thoughtfulness of the content.

Jen

Yeah. Well, this has been a lot to think about. [Doug laughs] It is! It's just, it's really kind of...

Doug

I've sort of been obsessed with it the past few months and it's really... there's a lot to it. There's a lot to unpack. I don't know how the future's going to unfold. Maybe this whole thing is going to go nowhere. Maybe browsers just decide, "No, we're not interested." And it only lives in script libraries and maybe those script libraries succeed or fail. I don't know. I hope that the future that we're building is one that includes annotations. I hope that that's the case. But I guarantee that no matter whether... if annotations does take off, I guarantee that if we had this, if we talked about this, say, five years from now, I would say, "Wow, I completely missed that. I totally did not get that that's what annotations was going to turn into." That the big service that's using annotations was gonna do X. It's like Minecraft, right? Oh, really? In an era where we have devices that you can hold in your hand that have photorealistic, immersive, interactive games, the one that all the 11-13 year old set are playing are things that look like pixelated games from, like Doom, right? From the early 90s. That's the one that's going to take off? Ok. That's the one that takes off. Ok, that's cool. I don't know what's going to happen with annotations. That's what I'm excited by. I can think of all sorts of things that might happen with annotations if we truly enable this. But I'm more looking forward to the things that I didn't see coming at all.

Jen

Yeah. Well, so people can follow you on Twitter. @shepazu is your name. Is that how you say it?

Doug

Yeah. It's the Japanese transliteration of my last name. So it's "shep-az". They would swallow the "u" so it's "shep-az". But I say "shepazu", that's... long story. [Laughs]

Jen

S-H-E-P-A-Z-U.

Doug

Yes.

Jen

And they can also read your blog, schepers.cc, which is S-C-H-E-P-E-R-S dot C C. But again, all of these things, including all of the coupon codes for the sponsors, are in the show notes at 5by5.tv. And they can, yeah, so read your blog. I should not just advocate for third party service following. [Both laugh]

Doug

Well, it would not take very much time to read my blog because I post so infrequently. But then again it wouldn't take much time to read my tweets either, because I don't tweet very frequently either. So, maybe I'm the wrong person to be talking about annotations. [Both laugh]

Jen

People can follow me on Twitter, @jensimmons or they can follow the show itself @thewebahead. Where I do not tweet very much because I don't want to spam people, I want people to follow the show and just get important stuff, like, "Hey, there's a new episode." So for any of you who ever wondered, "Hey, when is the next episode? Is there an episode coming up?" You can follow @thewebahead on Twitter and I will post that kind of news there. And, again, tell your friends about the show, subscribe, make sure you're subscribed and you could go drop a review or a rating in the iTunes store. That's a huge help, as well, to get the show visibility again, build back up our visibility. And listen next time. Hopefully there will be a show next week. Thank you everybody.

Doug

Thanks everybody.